During the planetary gear assembly, staking is a widely-used method for affixing pinion shafts onto the position. A reliable staking process not only prevents the movement of shaft during transmission operation, but also minimizes the distortion of the assembly due to the staking process. The quality of staking operations is determined by the component designs, the process parameters, and the staking tool geometry. It would be extremely time-consuming and tedious to evaluate these factors empirically; not even mention the requirement of prototypes in the early stage of a new program. A Finite Element methodology is developed to simulate the complete staking process including shaft press in, staking, and after staking tool release. The critical process parameters, such as staking force, staking length, shaft and holes interference amount, etc., are then evaluated systematically. Statistic tools are used to investigate the sensitivity and interactions between design and process parameters. The minimum force required to push shaft out of assembly position, an index of staking quality, is also calculated. A strong correlation between calculated and predicted press-in forces and minimum push out forces are found in this study.